Sure thing, Ski. We have an eval unit in our lab, and some of our colleagues in Europe really like them and want to standardize on them.
I like the interface and the simplicity, and I also like the compatibility and the reporting. It's really a purpose-built VMWare storage appliance. There are a few things that I wish they did better. For example, the dedupe doesn't extend to the disk tier -- only the SSDs. It would be more efficient if it deduped all of that data. I've heard (so this is entirely rumor) that the filesystem they are using isn't the most robust, but I can't find any supporting documentation for that besides the word-of-mouth of one person, so I'm not sure that's really accurate. It seems solid enough to me. I've also been told to test failover scenarios, because supposedly it can have trouble if a failover happens while it's already in a failed-over mode, but I haven't had a chance to test that yet. >From a functional perspective, I really like plugging it in and running with 13TB of datastore as a building-block. They are certainly getting accolades from the industry and are obviously a big player at VMWorld, too. -Adam On Fri, Aug 9, 2013 at 1:29 PM, Ski Kacoroski <[email protected]> wrote: > Ok, I have done a lot of research in the last few days on the Tintri and > am wondering if anyone else here has experience with it for server > virutalization (not VDI). From my research it really seems to be a > breakthrough technology as it is based on vm and vmdk, not on luns so they > can do all kinds of neat things that you cannot do with a traditional SAN > (map io/latency to a vmdk, replicate at vm level, etc.). So if you have > one, I would appreciate your comments on what you think about it. > > cheers, > > ski > > PS: Adam, thanks for mentioning them. > > > On 08/05/2013 09:35 AM, Ski Kacoroski wrote: > >> Adam, >> >> I have not looked at either of these options. My existing use is: >> >> - sata: 75% writes, 1600 iops with 3300 iops peak >> >> - hybrid: 49% writes, 2000 iops with 5000 iops peak >> >> I am looking for something that can grow my hybrid array and eventually >> in a year or two take over all the Equallogic arrays as they are EOL by >> Dell. >> >> cheers, >> >> ski >> >> >> On 08/05/2013 09:23 AM, Adam Levin wrote: >> >>> Are you firm on SAN? Have you looked at either Tintri or Tegile for a >>> NAS datastore? The Tintri in particular is extremely easy to set up. >>> The Tegile is a bit more powerful (and does both SAN and NAS, your >>> choice). They also dedupe, which is great for VMWare. They both have >>> full plugin support and are pretty nifty hybrid technology (some NL-SAS, >>> some SSD). >>> >>> What are your requirements for performance and capacity? >>> >>> -Adam >>> >>> >>> On Mon, Aug 5, 2013 at 12:16 PM, Michael Ryder <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Are you on ESXi5 or greater? >>> >>> One thing to consider is whether your chosen storage supports VAAI >>> >>> http://kb.vmware.com/**selfservice/microsites/search.** >>> do?language=en_US&cmd=**displayKC&externalId=1021976<http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=1021976> >>> >>> >>> I know that HP and EMC provide VAAI drivers for most of their gear, >>> across the past few years. >>> >>> You may want to review whether there are VAAI drivers for the Nimble >>> that you select: >>> >>> http://communities.vmware.com/**thread/436331?start=0&tstart=0<http://communities.vmware.com/thread/436331?start=0&tstart=0> >>> >>> Mike >>> >>> >>> On Mon, Aug 5, 2013 at 12:09 PM, Ski Kacoroski <[email protected] >>> <mailto:[email protected]>> wrote: >>> >>> Hi, >>> >>> I am looking for a new SAN for my 7 node, 200+ vm vmware >>> cluster. I currently have a Dell Equallogic SAN with one hybrid >>> node (completely full so I need to add on more fast storage >>> space) and 5 regular nodes of which 3 are going to be EOL by >>> Dell in a year (for no good reason). I have narrowed my choices >>> down to EMC (because I already have a VNX for NAS services), >>> Fujitsu, and Nimble Storage. >>> >>> From what I can see, the EMC and Fujitsu are your old school >>> style of SAN with lots of smaller disks to get the IOPs needed >>> and lots of flexibility (and more complexity) in setting them up >>> while Nimble is using software algorithms to get performance and >>> they focus on simplicity (kind of like the Dell). I am tempted >>> by the Nimble as I work in a very resource constrained >>> environment and simple is good, but I am a bit concerned about >>> how they will perform over a variety of workloads. >>> >>> So I appreciate your thoughts on Nimble vs traditional SAN's. >>> What workloads does a Nimble not do well on? Are there any >>> special features to a Fujitsu or EMC VNX that would make them >>> better with vmware? >>> >>> Thanks in advance. >>> >>> cheers, >>> >>> ski >>> >>> -- >>> "When we try to pick out anything by itself, we find it >>> connected to the entire universe" John Muir >>> >>> Chris "Ski" Kacoroski, [email protected] >>> <mailto:[email protected]>, 206-501-9803 <tel:206-501-9803> >>> or ski98033 on most IM services >>> ______________________________**___________________ >>> Discuss mailing list >>> [email protected] >>> <mailto:[email protected].**org<[email protected]> >>> > >>> >>> https://lists.lopsa.org/cgi-__**bin/mailman/listinfo/discuss<https://lists.lopsa.org/cgi-__bin/mailman/listinfo/discuss> >>> >>> <https://lists.lopsa.org/cgi-**bin/mailman/listinfo/discuss<https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss> >>> > >>> This list provided by the League of Professional System >>> Administrators >>> http://lopsa.org/ >>> >>> >>> >>> ______________________________**_________________ >>> Discuss mailing list >>> [email protected] >>> <mailto:[email protected].**org<[email protected]> >>> > >>> >>> https://lists.lopsa.org/cgi-**bin/mailman/listinfo/discuss<https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss> >>> This list provided by the League of Professional System >>> Administrators >>> http://lopsa.org/ >>> >>> >>> >> > -- > "When we try to pick out anything by itself, we find it > connected to the entire universe" John Muir > > Chris "Ski" Kacoroski, [email protected], 206-501-9803 > or ski98033 on most IM services > ______________________________**_________________ > Discuss mailing list > [email protected] > https://lists.lopsa.org/cgi-**bin/mailman/listinfo/discuss<https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss> > This list provided by the League of Professional System Administrators > http://lopsa.org/ >
_______________________________________________ Discuss mailing list [email protected] https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss This list provided by the League of Professional System Administrators http://lopsa.org/
